Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    words grouped with 'and' or 'or'

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     interieur
    -0.90
    اعة
    -0.88
    homePage
    -0.86
     tle
    -0.85
     ξ
    -0.85
     básica
    -0.84
    tarjeta
    -0.84
     peindre
    -0.83
    انية
    -0.82
    Annotated
    -0.82
    POSITIVE LOGITS
     outright
    1.23
     actual
    1.01
     sogar
    0.95
     downright
    0.95
     eventual
    0.91
     eventually
    0.89
     even
    0.87
    菲
    0.83
     insbesondere
    0.80
     short
    0.80
    Act Density 0.181%

    No Known Activations